What is the future of theory of mind, or TOM? That depends on what you mean by “TOM.” If by “TOM” you mean something people do – say, attribute mental states to others – the future looks bright. There is little chance people are going to stop doing those things anytime soon. But, if you mean a particular theoretical model of those things people do – say that they use something akin to a theory – then it is less clear what the answer is. Simulation theorists and others have challenged the (theory-)theory model for decades. Because the concept of TOM is ambiguous in this way (something people do, or a particular theoretical model thereof), however, you might think that concept is best avoided in future research on social cognition (see, e.g., Apperly Reference Apperly2011, p. 3; Goldman Reference Goldman2006, p. 10; Nichols & Stich Reference Nichols and Stich2003, p. 2).

When Veissière et al. raise the question of what the future of “thinking through other minds,” or TTOM, is, there is a similar ambiguity. Again, it all depends on what they mean by “TTOM.” Running through their paper is a fundamental conceptual muddle, as “TTOM” is defined – and the notion is employed – in two entirely different ways.

According to one definition, it is “a model of implicit cultural learning that we call ‘thinking through other minds’ (TTOM)” (sect. 1.3, para. 2). TTOM in this sense integrates other approaches, argues for compromises between externalist and internalist accounts, does (or does not) have certain ontological commitments, supports this or that theoretical view, provides mathematical formalizations, and so on. And of course, it is TTOM in this sense that is “a generic active inference (also known as FEP or variational) account of the acquisition of culture and mind-reading abilities” (sect. 5.1, para. 4).

According to another definition, however, TTOM refers to something people do – in particular, the forming of expectations about other people and their expectations. Thus, for example, “We call this intersubjective process of engaging others' expectations and inferences “thinking through other minds” (sect. 2.4, para. 8). TTOM in this sense consists of inferring others’ expectations, or learning to infer such things; individuals or groups of people may vary in terms of their capacities to “leverage” TTOM in this sense, and so on.

TTOM in this latter sense has a bright future, of course. Humans form habits and attitudes according to how they expect certain others to think and act. And they construct physical environments in such a way as to encourage certain behaviours (and discourage others), or make certain behaviourally relevant information salient, etc. A well-trodden path in the woods is a simple example; traffic lights, designed and positioned in such a way as to attract road users’ attention, a more complex one. None of this is likely to change, and Veissière et al. nicely bring out how important TTOM in this sense might be to solving the mystery of implicit cultural learning.

TTOM understood as a theoretical model might also fare well, although naturally the situation here is less clear. That model's proposed compromise between “internalist” and “externalist” approaches to social cognition is likely to meet resistance from hardliners on both sides. And even those who are sympathetic to the idea of such a compromise may quibble about the details.

The concept of TTOM, however, faces very uncertain prospects. A theoretical concept as fundamentally ambiguous as this is only fit to breed confusion. Thus, if Veissière et al. want to hold on to the concept of TTOM, it is crucial that they clear up the conceptual muddle sooner rather than later.